Evidence that a single replication fork proceeds from early to late replicating domains in the IgH locus in a non-B cell line

HAL Id: hal-01593094

https://hal-amu.archives-ouvertes.fr/hal-01593094

Submitted on 6 Dec 2018

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Distributed under a Creative Commons Attribution| 4.0 International License

Evidence that a single replication fork proceeds from early to late replicating domains in the IgH locus in a

non-B cell line

Olga V. Ermakova, Lam H. Nguyen, Randall D. Little, Christophe Chevillard, Roy Riblet, Nasrin Ashouian, Barbara K. Birshtein, Carl L. Schildkraut

To cite this version:

Olga V. Ermakova, Lam H. Nguyen, Randall D. Little, Christophe Chevillard, Roy Riblet, et al..

Evidence that a single replication fork proceeds from early to late replicating domains in the IgH locus

in a non-B cell line. Molecular Cell, Elsevier, 1999, 3 (3), pp.321-330. �10.1016/S1097-2765(00)80459-

1�. �hal-01593094�

HAL Id: hal-01593094

https://hal-amu.archives-ouvertes.fr/hal-01593094

Submitted on 6 Dec 2018

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Evidence that a single replication fork proceeds from early to late replicating domains in the IgH locus in a

non-B cell line

Ov Ermakova, Lh Nguyen, Rd Little, Christophe Chevillard, R Riblet, N Ashouian, Bk Birshtein, Cl Schildkraut

To cite this version:

Ov Ermakova, Lh Nguyen, Rd Little, Christophe Chevillard, R Riblet, et al.. Evidence that a single

replication fork proceeds from early to late replicating domains in the IgH locus in a non-B cell

line. Molecular Cell, Elsevier, 1999, 3 (3), pp.321-330. <10.1016/S1097-2765(00)80459-1>. <hal-

01593094>

Evidence that a Single Replication Fork

Proceeds from Early to Late Replicating Domains in the IgH Locus in a Non–B Cell Line

early replicating in cell lines where they are not ex- pressed (Hatton et al., 1988). It has been suggested that thesegenesreplicate earlyinSphase becausetheymay be located near an early replicating region (Goldman et al., 1984; Hatton et al., 1988; Simon and Cedar, 1996).

Olga V. Ermakova,* Lam H. Nguyen,*

Randall D. Little,* Christophe Chevillard, ^† Roy Riblet, ^† Nasrin Ashouian,*

Barbara K. Birshtein,* ^‡ and Carl L. Schildkraut* ^‡

*Department of Cell Biology Onalargerscale,mammalianchromosomesareorga-

Albert Einstein College of Medicine nized in early and late replicating regions that show New York, New York 10461 overall congruency with R and G bands, respectively

† Torrey Pines Institute for Molecular Studies (reviewed in Craig and Bickmore, 1993). Many of these San Diego, California 92121 regions have been observed by autoradiography to be replicated by clusters of replicons, which appear to be relatively synchronously activated during specific inter- vals early or late in the S phase (reviewed in Hand, 1978).

Summary Little is known, however, about how the transition from

early to late timing of DNA replication occurs. There are In non–B cell lines, like the murine erythroleukemia no examples in mammalian cell lines of a late replicating cell line (MEL), the most distal IgH constant region replicon located within an early replicating region nor gene,

^Ca

, replicates early in S; other heavy chain con- anearlyreplicatingrepliconwithinalatereplicatingclus- stant region genes, joining and diversity segments, ter. In addition, sharp transitions (with one possible ex- and the most proximal Vh gene replicate successively ception; Tenzen et al., 1997) between early and late later in S in a 3

⁹

to 5

⁹

direction proportional to their replicating regions have not been observed (Strehl et distance from

^Ca

. In MEL, replication forks detected al., 1997; Bilyeu and Chinault, 1998). One mechanism in the IgH locus also proceed in the same 3

⁹

to 5

⁹

to account for the progressive changes in replication direction for

^z

400 kb, beginning downstream of the timing between clusters of replicons activated in differ- IgH3

⁹

regulatoryregionandcontinuingtotheDregion, ent intervals of S phase would be the sequential activa- as well as within the

^Vh81X

gene. Downstream of the tion of replicons in the transition region. Another mecha- initiation region is an early replicating domain, and nism, for which we present evidence here, is a single upstream of

^Vh81X

is alate replicatingdomain. Hence, replication fork progressing from the last in a cluster of the gradual transition between early and late replicated early activated replicons to the first in a cluster of late domains can be acheived by a single replication fork.

activated replicons.

A unique pattern of replication timing is observed (see Figure 1) for the immunoglobulin heavy chain gene (IgH)

Introduction locus (and to some extent the Ig kappa light chain locus;

Hatton and Schildkraut, 1990). In non–B cells, like the The time during S phase in which different replicons are murineerythroleukemiacellline(MEL),replicationtiming replicated is regulated. In general, loci that are ex- of specific segments of the IgH locus is linearly propor- pressed in a tissue-specific manner replicate during the tional to their map locations: sequences 3

⁹

of the locus first third of S in cells in which they are expressed but replicate very early during S phase in non–B cells (Mi- often replicate later in S in cells in which they are not chaelson et al., 1997), and replication of successive IgH expressed (see e.g., Goldman et al., 1984; Hatton et al., segments, that is, Ch genes and Jh and Dh segments, 1988;SimonandCedar,1996).Forexample,the

^b

-globin becomes progressively later in S phase (Brown et al., gene cluster replicates early in the human K562 cell line 1987).Allofthevariableregionsegmentswehaveexam- in which it is expressed, but late in S in HeLa cells in ined—occupying at least 700 kb at the 5

⁹

side of the which it is not expressed (Dhar et al., 1988): regardless IgH locus—replicate during the second half of S (Calza of expression, the same initiation site is used (Aladjem et al., 1984). Replication timing is different in cells of the etal.,1995).AsimilarpatternoftimingofDNAreplication B lineage,in whichDNA rearrangementsand expression has been observed for genes for

^a

-fetoprotein and albu- of antibody genes exclusively occur. In both mature B min, the T cell receptor

^b

chain constant regions and cells and pre–B cells, sequences immediately 3

⁹

of the the cystic fibrosis transductance regulator (Goldman et IgH locus, the entire IgH coding locus, and the ex- al., 1984; Hatton et al., 1988; Simon and Cedar, 1996). pressed Vh genes all replicate early in S phase (Brown Each replicates early in S in cells in which they are etal.,1987;Michaelsonetal.,1997;O.V.E.,unpublished expressed and late in cell lines in which they are silent. data). In pre–B cells, early replication also extends to There are certain interesting exceptions to this observa- the entire Vh gene locus (Hatton et al., 1988).

tion. Forexample, the mouse

^a

-globin clusterand genes Tobegin tounderstand themolecular mechanismthat coding for complement proteins C4 and factor B are regulates the timing of DNA replication in the immuno- globulinheavychainconstantregiongene(IgH-C)locus, we examined the direction of replication fork movement

‡ To whom correspondence should be addressed (e-mail: schildkr@

aecom.yu.edu [C. L. S.] and [email protected] [B. K. B.]). through this locus in MEL (non–B) cells by neutral/alka-

Figure 1. The Immunoglobulin Heavy Chain Region Links Early and Late Replicated Zones of the IgH Locus

(A) Map of the IgH locus in non–B cells (MEL).

The germline configuration of the IgH locus on murine chromosome 12q32 is shown (Shimizu et al., 1982). The entire locus is ap- proximately 3 Mb in size and contains more than 100 variable region genes (Vh) organized in fifteen families, four joining segments (Jh), twelve diversity segments (Dh), and eight constant region genes (Ch), represented by filled squares. The 3

⁹

RR and the array of Ch, Dh, and Jh gene segments occupies

z

400 kb and is referred to as the IgH-C locus. Only two variable gene families, VhS107 and Vh7183, are shown above.

^Vh81X

is the most 3

⁹

member of the most proximal variable re- gion gene family (Vh7183), all members of which are located in an interval of 100–150 kb. The E

^m

enhancer is located between the Jh segments and

^Cm

gene (Gillies et al., 1983;

Friedman and Brewer, 1995). Downstream of the

Ca

gene, four enhancers, represented by shaded squares, comprise the 3

⁹

RR (Saleque et al., 1997). In non–B cells, such as MEL, the locus is unrearranged and transcriptionally inactive. The 3

⁹

RR is in an inactive chromatin configuration and does not display any DNase I hypersensitivity (reviewed in Bir- shtein et al., 1997). The scale in kb is shown below the map. The position of the

^Ca

gene is indicated at zero. The left to right, 5

^9→

3

⁹

, orientation of the IgH locus is with respect to transcription of IgH genes in B cells.

(B) Time in S when replication of the murine IgH locus occurs in MEL (non–B) cells. The y axis indicates the nuclear DNA content (C values) at which IgH segments replicate as a function of their distance in kb (x axis) from the

Ca

gene. C is the haploid DNA content of the cells at the time in S phase when a particular DNA segment replicates. For example, at the G1/S boundary the C value is 2.0, and after DNA replication is completed, the C value is 4.0. The

^V1

gene, a member of the VhS107 family, is located 200–250 kb 5

⁹

of

^Vh81X

. The other three members of this family are spread over a region of at least 200 kb further upstream.

The

^Vh3609N

gene is about 700 kb upstream from

^Vh81X

. Some of the data (open circles) presented here have been published previously (Brown et al., 1987; Michaelson et al., 1997). Replication timing was carried out as previously described (Brown et al., 1987). Probe positions (determined to within at least

⁶

50 kb) and timing data for segments located at

^Ca2

300 and

^Ca2

220 as well as positions for segments upstream of

^Vh81X

are newly obtained.

(C) Models for the replication of the IgH-C locus in non–B cell lines. Each small bubble represents a single bidirectional replicon. Larger bubbles represent regions where forks from two or more replicons have converged. (1) Only one bidirectional replicon is activated near the beginning of S phase, and a single replication fork progresses in the direction of the Vh region gene families at a rate of about 1.5 kb/min.

Constant region genes replicate during the first half of the S phase, D region genes in the middle S phase, and V region genes replicate during the second half of S. (2) Several bidirectional replicons are activated sequentially in the S phase in the IgH locus. Positions in this figure are approximately as shown on the map in (A).

line (N/A) two dimensional (2D) gel electrophoresis (Na- Results wotka and Huberman, 1988). Together with the results

of neutral/neutral (N/N) 2D gel electrophoresis (Brewer The IgH-C Locus Represents a Gradual Transition and Fangman, 1987; Friedman and Brewer, 1995), the between Early and Late Replicating Regions

data presented here are consistent with a model of a Recent studies have generated a complete contig of the

single replication fork that replicates IgH-C and may IgH locus in overlapping yeast and bacterial artificial

progress as far as 400 kb. These studies have enabled chromosomes (YACs, BACs) (C. C. and R. R., unpub-

us to identify a putative initiation region(s) from which lished data), providing the basis for a complete map of

replication forks originate. In MEL cells, the IgH-C locus the locus (Figure 1A), determination of the distances

represents a gradual transition between early and late between the genes, and additional probes, all of which

replicated zones. It remains to be determined whether were essential for the further studies of DNA replication

a single replicon represents a general method for pro- in this locus presented here. Utilizing these newly identi-

ducing the transition between early and late replicating fied probes for sequences both 3

⁹

of and throughout

the IgH locus, we were able to assess the relationship

temporal domains of mammalian chromosomes.

distance from

Ca

. Importantly, 1B. between

The

Ca

distances gene and

replicates timing, near as

the summarized beginning of in

S, Figure and in replication

downstream proportion along of to

Ca

the the (st

locus becomes progressively later arting at about

^Ca2

80, that is, 80 kb downstream of

^Ca

), the temporal pattern of DNA replication is not proportional with respect to distance.

Both

Ca2

220 and

Ca2

300 replicate very early in S:

Ca2

220 at C

⁵

2.28 and

^Ca2

300 at C

⁵

2.3. Hence, a region of at least 250 kb, as assessed by four sequences downstream of

^Ca

(

^Ca2

54,

²

80,

²

220, and

²

300) repli- cates very early in S phase.

In the variable region, we find that

^Vh81X

, which is the most 3

⁹

member of the Vh7183 family and is located 20–60 kb 5

⁹

of the most 5

⁹

Dh gene, replicates at 3.3 C, with the same linear relationship of distance from

Ca

to time (Figure 1A). However, this relationship is no longer evident for other variable region genes, including the other six members of the Vh7183 family, the four mem- bers of the VhS107 family, and the most distal gene analyzed,

^Vh3609N

gene (which is located

^z

700 kb 5

⁹

of

^Vh81X

); all replicate in approximately the same inter-

val late in S phase, from 3.3 C to 3.6 C. Together, these data indicate that in MEL, the timing Figure 2. Neutral/Alkaline (N/A) 2D Gel Electrophoresis Detects the Direction of Replication Fork Movement and Site of Initiation of DNA

of DNA replication is regulated differently for the IgH Replication

locus and its flanking sequences. A region of at least In the upper half of the figure are schematic representations of the different replicative intermediates that are generated in three

300 kb downstream of

Ca

is early replicating, while the

adjacent restriction fragments. The origin of bidirectional replication

Vh genes upstream of the constant region are late repli- is located in the restriction fragment in the center, and forks proceed

cating. Between these early and late replicating regions, outward from the origin in both directions. Unbroken lines represent

sequences extending over a 400 kb region from

^Ca2

80 parental strands, and dotted lines represent nascent strands. Below

through the Ch, Jh, and Dh gene segments to the first the replicatingstrands, black rectanglesindicate the locationsof the

Vh gene,

^Vh81X

, replicate progressively later in S phase. probes used for the N/A 2D gel electrophoresis. The lower portion of the figure shows the Southern transfer hybridization patterns expected from each probe. In the first dimension, DNA is separated

Models for the Replication of IgH-C primarily according to mass. The second dimension is performed at alkaline pH so that nascent strands separate from parental

Constant Region Genes

strands and migrate according to their size, as shown at the bottom

Replication timing data for IgH sequences in MEL cells of the figure (Nawotka and Huberman, 1988). The 1X spot is indi-

could best be explained by two different models (shown cated and represents the linear molecules of the same size as the

in Figure 1C). The first (Figure 1C [1]) involves activation restriction fragment. The direction of replication fork movement is

of a single replicon that is located in the earliest replicat- revealed by hybridization to probes located at different positions

ingregiondownstreamof

Ca

.Onereplicationforkwould relative to the ends of the restriction fragments. If forks move from the right to the left end of the restriction fragment (shown on the

then proceed through the eight constant region genes,

left part of the figure), the probe from the right end of the restriction

J and D regions, and finally the most 3

⁹

heavy chain fragment detects nascent strands of all sizes. The probe from the

variable gene (

^Vh81X

). IgH sequences in this interval left end will detect only long nascent strands. The probe from the

would comprise a single replicon. The second model middle of the restriction fragment detects only nascent strands that

(Figure 1C [2]) involves the sequential activation of more are not less than half the size of the restriction fragment. If the forks

than one bidirectional replicon in the IgH-C locus during move through the restriction fragment from left to right (shown on the right part of the figure), the probe from the left end detects

the S phase. These models can be distinguished by

nascent strands of all sizes and the probe from the right end detects

analysis of the direction of replication fork progression only long nascent strands. The restriction fragment in the middle

as assessed by N/A 2D gel electrophoresis (Nawotka contains an origin of bidirectional replication, so that forks move

and Huberman, 1988). If replication forks originate from from the center to both left and right of the restriction fragment. In

a single downstream origin, forks will progress in only this instance, probes from either end of the restriction fragment

one direction through the IgH-C locus (Figure 1C [1]). detect only nascent strands, which are not smaller than half of the size of the restriction fragment. The probe in the middle of the seg-

In contrast, replication forks from multiple bidirectional

ment detects nascent strands of all sizes that originate from the

replicons activated sequentially would progress out- initiation site.

ward in both directions.

Direction of Replication Fork Movement through

the IgH-C Locus, Revealed by N/A 2D replication intermediates was crucial prior to electro- phoresis(seeExperimentalProcedures).First,MELcells Gel Electrophoresis

Due to the complexity of the mammalian genome and from an exponentially growing culture were fractionated by size by centrifugal elutriation (Brown et al., 1987), the extremely low abundance of molecules actively rep-

licating single copy loci, a series of steps to enrich for and cells in early, early middle, and middle late fractions

Figure 3. Replication Forks Originate Downstream of the 3

⁹

RR and Progress in the Direction of the Vh Region in MEL Cells

Restriction fragments (size is indicated below each fragment) examined by N/A 2D gel electrophoresis are shown below the map. The position of the probes at 3

⁹

or 5

⁹

ends of the restriction fragments are indicated by the solid rectangles, and the probes are summarized in Table 1.

Autoradiograms are shown under each restriction fragment. DNA was obtained from cells in the early and early middle stages of S phase, except for the analysis of the 3.5 kb PvuII/EcoRI fragment (DNA prepared from exponentially growing cells) and for the

Vh81X

segment (DNA prepared from the middle portion of S phase). In other experiments (data not shown), the 3.5 kb PvuII/EcoRI fragment was used to analyze DNA obtained from cells in the early and early middle stages of S; and

^g3

region and

^m

region segments have been examined using DNA prepared from exponentially growing cells. In each case, we observed the same results. Each 3

⁹

end probe detects nascent strands of all sizes, while 5

⁹

end probes detect only long nascent strands. This indicates that replication forks originate downstream of the 3

⁹

RR and continue to progress in the 3

⁹

to 5

⁹

direction. For several segments (

^e

region,

^g2a

region, and

^g2b

region), after hybridization using the 5

⁹

end probe, the transfer was stripped and a 3

⁹

end probe was used. Note that probes 7 and 8, as well as 13 and 14, are located in close proximity to each other flanking an EcoRI site. In these two regions (

^g2b

and

^Vh81X

), the 3

⁹

and 5

⁹

end probes detect two neighboring segments rather than the same segment. Thus, for these two pairs, the pattern is reversed, and the small nascent strands are observed in the autoradiogram on the left. Arrows under autoradiograms indicate the direction of replication fork movement. E, EcoRI; B, BamHI; P, PvuII.

of S phase were collected separately. DNA associated As shown in Figure 3, we have used twelve probes (Table 1) to examine seven segments of IgH-C se- with the nuclear matrix was then isolated and applied

to a sucrose gradient. DNA devoid of high molecular quences. A probe for the 3

⁹

end of each segment de- tected nascent strands ranging from

^z

300 bp to the full weight fractions was then separated by benzoyl naph-

thyl DEAE– (BND-) cellulose chromatography. Figure 2 length of the linear segment while a probe for the 5

⁹

end detected only nascent strands of large sizes. The schematically represents how N/A 2D gel electrophore-

sis reveals replication fork movement from a bidirec- exclusiveidentificationoflargereplicativeintermediates at the 5

⁹

end of the segment is not due to the preferential tional origin of DNA replication.

Figure 3 shows the direction of replication fork move- loss of small products during DNA preparation because these small products can be detected at the 3

⁹

end of ment through IgH sequences in MEL, as examined by

N/A 2D gel electrophoresis. For example, the most 3

⁹

the same segment on the same DNA transfer (Figure 3).

These results show that replication forks progress from segment of this region, a 3.5 kb PvuII/EcoRI fragment

in the 3

⁹

RR (extreme right in Figure 3), has been exam- the 3

⁹

end to the 5

⁹

end of each of the seven segments.

To further extend these studies, we have also per- ined as follows. After N/A 2D gel electrophoresis, DNA

from the gel was transferred to a membrane and hybrid- formed N/A 2D gel electrophoresis on two segments containing the

^Vh81X

gene and have found that replica- ized to probes 1 and 2 for either end of the 3.5 kb

PvuII/EcoRI segment (

^Ca2

45) on two different blots, or tion forks also progress in the 3

⁹

to 5

⁹

direction (Figure 3). Since the

^Vh81X

gene replicated later in S phase, we sequentially with intermediate stripping. Linear mole-

cules of the size of the restriction segment (1X spot, see examined replication intermediates from exponentially growing cells as well as from cells in the middle part of Figure 2) are detected at the upper right part of the

autoradiogram (shown below the segment). The hori- S phase. Together with studies on timing (Figure 1B), whichindicatethat

^Vh81X

replicateswiththesamelinear zontal line emanating from the 1X spot represents the

parental strands. The vertical line that originates from relationship as do the Ch genes, this result suggests that a single bidirectional replicon replicates Ch, Jh, the 1X spot reflects nicked molecules and is generally

indicative of the quality of the DNA preparation. Nascent Dh, and the

^Vh81X

gene, although we cannot exclude activation of an additional replicon that is located in the strands are found along a diagonal line emanating from

parental strand molecules. The probe from the 3

⁹

end Dh region. These results are most consistent with a single bidirectional replicon with an initiation region that of the fragment detects nascent strands of all sizes. As

seen on the adjacent autoradiogram, the probe from is located downstream of

^Ca2

45.

Several important controls have been carried out to the 5

⁹

end of the segment detects only large nascent

strands. These results show that replication forks move validate the 2D gel techniques used here. The majority oftheN/Agels(andmanyoftheN/N2Dgels[seebelow]) in the 3

⁹

to 5

⁹

direction through the 3.5 kb PvuII/EcoRI

segment, implying that they originate from a region lo- were performed two or more times using independent matrix preparations with the same result. In addition, cated further 3

⁹

, that is, downstream of this segment.

These data are consistent with the timing data shown weshowedthat,asexpectedinN/A2Dgelelectrophore-

sis, a probe from the middle of a segment detected

in Figure 1B.

Table 1. Probes from the IgH Locus that Have Been Used

Probe Number Probe Preparation References

1 0.7 kb HincII/EcoRI fragment, located near the 3

⁹

-most EcoRI site in 3

⁹

RR (Michaelson et al., 1997) 2 PstI-2.3 is a 2.3 kb PstI/PstI fragment; contains an internal PvuII site that cuts the (Gregor and Morrison, 1986;

probe into 1.6 kb and 0.7 kb segments. The 1.6 kb part of the probe is located Michaelson et al., 1995) at the 5

⁹

EcoRI/PvuII segment.

3 2.0 kb XbaI/BamHI probe has been excised from a 4.6 kb BamHI/BamHI segment (Calvo et al., 1991) (p4.6C

^e

) containing

^e

constant region gene coding sequence; located at 3

⁹

end of the BamHI/BamHI fragment.

4 0.3 kb BamHI/KpnI probe, that is located at the 5

⁹

end of the 4.6 kb BamHI/BamHI (Calvo et al., 1991) (p4.6C

^e

) segment containing

^e

constant region gene coding sequence.

5 The 0.8 kb XbaI/EcoRI probe, located at the 3

⁹

end of the 4.2 kb EcoRI segment (Eckhardt and Birshtein, 1985) (pS

^g

2a-3), encompassing the switch region for the

Cg2a

gene.

6 The 1.0 kb EcoRI/SacI probe located at the 5

⁹

end of the 4.2 kb EcoRI segment (Michaelson et al., 1995) (pS

^g

2a-3), encompassing the switch region for the

Cg2a

gene.

7 The EcoRI/BglII probe is located upstream of the

^g2b

gene coding sequence. The 1.9 (Roeder et al., 1981;

kb probe is derived from the 6.8 kb p

^g

2bR1.4 subclone and is located at the 5

⁹

Lang et al., 1982) end of the EcoRI segment.

8 pBR1.4 probe is located at the 3

⁹

end of the 6.0 kb EcoRI segment that encompasses (Roeder et al., 1981;

the

^g2b

switch region. Lang et al., 1982)

9 The 1.9 kb HindIII/BamHI probe at the 3

⁹

end of the 6.0 kb BamHI/BamHI fragment (Lang et al., 1982) (p

^g

3-6/0), which contains 2.0 kb

^Cg3

gene.

10 The 2.0 kb BamHI/KpnI probe, which encompasses the

^Cg3

protein coding sequence, (Lang et al., 1982) is located at the 5

⁹

end of the 6.0 kb BamHI/BamHI segment (p

^g

3-6/0).

11 The probe from the 3

⁹

end of the 4.5 kb EcoRI/BamHI segment (J-14) is a 2.0 kb (Lang et al., 1982) HindIII/BamHI fragment, which contains exon CH1 and part of exon CH2 of the

Cm

gene. In MEL cells, an EcoRI/BamHI segment (J-14) is 5.8 kb.

12 The 5

⁹

end probe is a 900 bp EcoRI/SacI fragment located at the 5

⁹

end of the 4.5 kb (Lang et al., 1982) EcoRI/BamHI segment (J-14); the SacI site is located 3.3 kb from exon CH1 of the

Cm

gene. In MEL cells, this EcoRI/BamHI segment (J-14) is 5.8 kb.

13 0.8 kb AseI fragment containing the 3

⁹

half of the

^Vh81X

gene from pVh81X (Yancopoulos et al., 1984)

14 0.9 kb DraI fragment from pVh81X (Yancopoulos et al., 1984)

15 The probe is 2.3 kb located at

^Ca

-80. (Michaelson et al., 1997)

16 The hs 4 enhancer probe (Michaelson et al., 1995)

17 The hs3 enhancer probe is 1.2 kb XbaI/Sau3A. (Saleque et al., 1997)

18 J-11 probe is 2.0 kb BamHI/EcoRI fragment (Lang et al., 1982)

19 1.6 kb probe prepared from BAC395D22 is located at the 3

⁹

end of the HindIII segment. (Sidow et al., 1997;

present study) 20 A 1.8 kb NotI fragment containing the 3

⁹

end of the Serrate gene (Lan et al., 1997)

21 The 500 bp probe from the 3

⁹

end (Sp6 end) of BAC 397G20 (present study)

22 1.0 kb probe is from the 3

⁹

end of BAC 225H9. (present study)

23 A 1 kb EcoRI/BamHI fragment of the

^Vh3609N

gene (Tutter et al., 1991

nascent strands of intermediate sizes and not the small- the IgH locus, and we cannot exclude the possibility that there are additional smaller replicons in the regions est sizes (data not shown). Furthermore, there is no

technical limitation that preferentially results in the de- not yet examined. In addition, in a few instances, we have used the same probe to detect the 3

⁹

end of one tection of forks proceeding exclusively in the 3

⁹

to 5

⁹

direction. Based on studies of the murine

^b

-globin locus segment as well as the 5

⁹

end of an overlapping seg- ment. In a few instances, we have also used a probe at in which an initiation region has been mapped by three

differenttechniques(M.Aladjemetal.,personalcommu- the 3

⁹

end of one segment and the 5

⁹

end of an adjacent segment. Based on timing data, putative smaller repli- nication; L. H. N., O. V. E., and C. L. S, unpublished

data),weexpectedreplicationforkstoproceedpredom- conswould havetobe activatedsequentiallyin Sphase.

Moreover, if there are additional bidirectional replicons inantly in a 5

⁹

to 3

⁹

direction (opposite to IgH) when

probing the 3

⁹

side of an initiation region. When the in the locus, itseems likely that replicative intermediates other than those containing a single replication fork (i.e., same DNA transfers used to study the IgH region were

probed for the

^b

-globin locus, we detected these pre- molecules containing replication bubbles or converging replication forks) would be detected in the studies using dicted forks (L. H. N., O. V. E., and C. L. S., unpublished

data). N/N 2D gel electrophoresis, as described below.

While we have been able to obtain probes spanning

the locus, unique probes are limited in number due to Shape of Replicative Intermediates in the IgH Locus Revealed by N/N 2D Gel Electrophoresis

the unusual repetitive content of this locus. In addition

to a relatively high content of Line1 elements (Herring We examined ten segments between the

^Ca

gene and the J region, and we have only detected replication et al., 1998), the IgH locus is comprised of families of

related and cross-hybridizing elements, that is, the Ch, intermediates that migrate as molecules containing a single replication fork (Figure 4). Importantly, we did not Jh, Dh, and Vh gene families. Therefore, we have not

been able to examine overlapping segments throughout detect initiation and termination events, an observation

Figure 4. Initiation and Termination Events Have Not Been Detected in MEL by N/N 2D Gel Electrophoresis

The relative positions of the restriction fragments that have been examined are indicated, and the probes within restriction fragments are shown as black rectangles. A description of the probes (see numbers) is presented in Table 1. Like N/A 2D gel electrophoresis, the N/N 2D gels also separate the DNA molecules according to their size in the first dimension. In the second dimension, the gel conditions maximize the effect of the shape of DNA molecules on their migration. This technique can detect replication intermediates that correspond to initiation events, and to termination and to single replication forks proceeding through the segment (Brewer and Fangman, 1987; Friedman and Brewer, 1995). Only replication intermediates in which forks progress from one end of the segment to the other end were detected in the IgH locus.

E, EcoRI; B, BamHI; H, HindIII; P, PvuII; S, SacI.

that argues against the presence of bidirectional se- (probe 21) and one segment at

^Ca2

300 (probes 20 and 19) were used to determine replication fork direction quentially activated replicons in the IgH-C locus. As a

control for these studies, using nuclear matrix-associ- (Figure 5). Nascent strands of all sizes were detected in all instances, indicating that replication forks progress ated MEL cell DNA and N/N 2D gel electrophoresis with

the same techniques as in the present study, we were in both directions. In contrast, replication forks progress exclusively in the 3

⁹

to 5

⁹

direction through

^Ca2

45 (Fig- able to detect replicative intermediates containing bub-

bles indicative of a replication origin in the mouse rDNA ure 3). Thus, the direction of replication fork movement changes within a region of

z

55 kb, in the interval be- locus (P. Galgano and C. L. S., unpublished data). Thus,

the absence of detectable replication intermediates tween

^Ca2

45 and

^Ca2

100, consistent with the location of an initiation region within this interval. Therefore, we containing replication bubbles in the IgH locus is not

dueto alimitation ofthetechnique but,rather, ischarac- have identified the regionfrom which the replication fork originates that makes the transition between the early teristic of the locus itself.

Other investigators used an estimation of the abun- andlatereplicatingboundariesoftheIgHlocus.Further- more, at least one additional initiation region must be dance of nascent strands to identify a specific, relatively

weak origin of DNA replication within 500 bp of the located downstream of

^Ca2

300, although other initia- tion regions may also be present in the interval between intronic enhancer, E

^m

, in both B cells and in fibroblasts

(Ariizumi et al., 1993). Neither of the 2D gel electrophore-

^Ca2

100 and

^Ca2

300. Interestingly, as previously shown (Figure 1B), the replication timing of sequences down- sis techniques used here detected this origin. N/A 2D

gel electrophoresis did not identify any replication forks stream of

^Ca2

80 no longer changes with distance.

that would originate from an E

^m

-containing DNA seg- ment. Nor did we detect this origin by N/N 2D gel analy-

sis,evenwhenweexaminedanE

^m

DNAsegment(Figure Discussion 4, BamHI/SacI segment) in which the putative origin of

DNA replication is located within the middle third of the Examination of DNA replication of the IgH gene cluster in MEL, based on previous data and data presented in examined segment, where initiation is best detected.

Thus, if thisE

^m

-associated origin is present inMEL cells, this report, has shown that: (1) DNA segments between

Ca2

300 and

Ca2

80 replicated early during S phase; (2) it must be weak.

segments within a 400 kb interval including Ch, Jh, Dh, and the first variable region gene,

^Vh81X

, replicate pro- Direction of Replication Fork Progression Changes

in the Early Replicating Region Located gressivelylaterindirectproportiontotheirdistancefrom the putative earliest replicating sequences; and (3) other Downstream of the 3

⁹

RR

Probes located at the 3

⁹

end and at the 5

⁹

end of two variable region gene segments detected replicate late

in S. These studies indicate that the IgH replicon is

overlapping segments at

^Ca2

100 (probe 22) and

^Ca2

220

Figure 5. Direction of the Replication Fork Movement in the Region Downstream of the 3

⁹

RR

The size of examined fragments and their ap- proximate distance from the

^Ca

gene, as ob- tained by mapping of BAC clones, are indi- cated. Probes utilized to assay the direction of replication fork movement are indicated by black rectangles. The autoradiograms of 2D gel electrophoresis are presented under the fragments. The direction of replication fork movement,asdeterminedbyN/A2Dgelelec- trophoresis, is indicated by arrows at the top ofthe figure.Two overlappingsegments have been examined at

Ca2

100. The probe (probe 22) from the 3

⁹

end of one segment, which is also at the 5

⁹

endof the overlapping segment, detects nascent strands of all sizes. This re- sult demonstrates that the replication forks move in both directions through this seg- ment. A putative initiation region has been denoted in the region where the change in direction of replication forks occurs. For the segments at

^Ca2

220, similar results were obtained using two different DNA transfers from independent experiments. Probes from both ends of the restriction fragment at

^Ca2

300 (probes 20 and 19) also detect nascent strands of all sizes, indicating that replication forks progress in both directions through this segment. E, EcoRI; B, BamHI; H, HindIII.

bounded on one side by an early replicating region and studieshaveestimatedthatonly5%ofmammalianrepli- cons could be unidirectional (Hand, 1975). Although uni- a putative initiation region located 3

⁹

of the 3

⁹

RR and

on the other side by late replicating Vh region genes. directional replication has been described for plasmid replication of prokaryotes, mitochondrial DNA, and some We performed N/A 2D gel electrophoresis on seven

segments within the 400 kb region that encompasses viral genomes (Kornberg and Baker, 1992), it has not been reported in yeast or for the few replicons studied the IgH-C locus and two restriction segments in the

vicinity of the nearest variable gene,

^Vh81X

. We con- on the molecular level in mammalian cells. In addition, replicons in mammalian cells that are located in proxim- clude that replication forks progress exclusively in the

3

⁹

to 5

⁹

direction; however, we would not detect less ity to each other are usually activated simultaneously, but not sequentially (Huberman and Riggs, 1966; re- than about 10% of replication forks if they progressed

intheoppositedirection.Whilethelackofuniqueprobes viewed in Hand, 1975). Therefore, we conclude that our data are most consistent with a single replication fork preventedexaminationofa

z

100kbregionthatcontains

the array of Dh gene segments, our studies show that proceeding through the IgH-C locus in a 3

⁹

to 5

⁹

direc- the replication time of the

^Vh81X

gene is consistent with tion. This replicon is activated near the beginning of S the same linear relationship between replication time phase in cells, like MEL, in which the IgH locus is not and distance from the

Ca

gene as observed for all the expressed. Evidence for relatively large replicons has Ch, Jh, and Dh segments. Additionally, N/A 2D gels been previously presented (e.g., Yurov et al., 1977; re- indicate that replication forks progress through the seg- viewed in Hand, 1978).

ment containing the

^Vh81X

gene in the same direction as observed for the other IgH segments that we have

studied. These data strongly suggest that about 400 kb Localization of Initiation Events

Replication forks progress in both directions in se- of the IgH gene cluster is replicated by a single replicon,

which is activated at the beginning of S phase and prog- quences located at

Ca2

100,

Ca2

220, and

Ca2

300 downstream of the IgH-C locus (Figure 5). These data resses until a substantially late time (C

⁵

3.4) in the S

phase (Figure 6). suggest the presence of an initiation region or a se-

quence-specific bidirectional replication origin in the in- We cannot totally rule out the possibility of multiple,

sequentially activated unidirectional replicons by these terval between

^Ca2

45 and

^Ca2

100 (for a schematic representation, see Figure 6). These data also provide experiments. Except at potential initiation sites, the N/A

2D gel electrophoresis patterns would be identical for a evidenceforatleastoneadditionalreplicon,whichorigi- nates downstream of

^Ca2

300. Other replicons may be singlebidirectionalrepliconandformultiplesequentially

activated unidirectional replicons. In addition, N/N 2D located between

^Ca2

100 and

^Ca2

300. If discrete initia- tionsitesare present,weshoulddetect segmentswhere gel electrophoresis should detect initiation or termina-

tion events in particular fragments, and our experiments replication forks progress predominantly in one direc- tion. Indeed, at one location (

^Ca2

120), the majority of detected neither. It remains possible, however, that if

initiation or termination occurred in different segments replication forks proceeds in one direction (data not shown). Alternatively, the identification of replication indifferentcellpopulations,theirfrequencyatanysingle

location could be so low as to preclude detection. forks moving in both directions could be explained by an initiation zone. Forks progressing in both directions On the other hand, there is limited evidence for unidi-

rectionalrepliconsinmammaliancells.Autoradiography have previously been detected in rDNA (Little et al.,

Figure 6. Summary of Replication Fork Movement in the Murine IgH Locus in MEL (Non–B) Cells

The structural organization of the locus is shown in the upper part of the figure. Seg- ments that have been assayed by N/A 2D gel electrophoresis are shown by pairs of arrows (schematically represented in the middle part of the figure); each arrow represents the loca- tion of the probe. These data are consistent with a single replication fork progressing in only one direction through the IgH-C locus from 3

⁹

of 3

⁹

RR to 5

⁹

(

Vh81X

) for

z

400 kb.

Replicationforksoriginatedownstreamofthe 3

⁹

RR, within an

z

55 kb region (

Ca2

100 to

Ca2

45) where we detected a switch in the direction of replication fork progression. Fur- ther 3

⁹

, replication forks move in both direc- tions, indicating the presence of a second replicon in the region located at least 300 kb downstream of

^Ca

. The bubbles represent either site- specific initiation or a delocalized initiation zone. The IgH-C region, Jh, Dh, and first variable gene (

^Vh81X

) represent a transition between early and late replicated regions (shown at the bottom). A single replication fork apparently originating from the early activated replicon travels for almost the entire S phase to the late replicated region. The gradient indicates the gradual change in replication timing of the IgH-C. Early and late replication is achieved by clusters of replicons activated at similar times in S. The bottom diagram is not aligned with the others in the figure.

1993) and in the DHFR initiation zones (Dijkwel et al., of the direction of replication fork progression between early and late replicating DNA sequences (Figure 6).

1994).

Further detailed studies are in progress using N/N 2D The IgH locus could represent a model for many other similar transition regions that are present in mammalian gels to detect initiation events and to localize initiation

site(s) in the IgH gene cluster. It will then be of interest genomes. For example, timing of DNA replication was shown to change gradually at a R/G band boundary on to compare the IgH-associated initiation regions with

others that have been intensively studied, that is, dihy- chromosome 13 (between 13q14.3/q21.1) over a dis- tance of 500 kb (Strehl et al., 1997), an observation drofolate reductase (DHFR) in Chinese hamster ovary

cells (for review see Dijkwel and Hamlin, 1996; Kobay- originally interpreted to result from progressively later activation of additional origins of DNA replication. A ashi et al., 1998), ribosomal RNA genes (Little et al.,

1993; Yoon et al., 1995; Gencheva et al., 1996), and the similar gradual transition has been observed between cytogenetic band boundaries on the human X chromo-

b

-globin gene locus in human cells (Kitsberg et al., 1993;

Aladjem et al., 1995, 1998). some (Bilyeu and Chinault, 1998). Another transition in DNA replication timing has been identified within a 100 kb boundary between R and G bands located between classes II and III in the human major histocompatibility The IgH Locus as a Model for Understanding

Transitions between Other Early and Late complex (Tenzen et al., 1997). Each of these observa- tions could be accounted for by a single replication fork Replicated Chromosomal Regions

In the present study, we provide evidence that multiple that accomplished the transition.

replicons are activated in the early replicated region downstream of

Ca

. Furthermore, our observation that

Vh genes at the other end of the IgH locus that are Evidence for a Developmentally Regulated Origin Replication of the IgH locus is very different in cell lines located several hundred kb from each other replicate

within a narrow interval late in S phase also suggests of the B lineage in which IgH genes are expressed. For example, in pre–B cell lines, the entire IgH locus—from the presence of several replicons. Connecting these

early and late replicating domains is the IgH-C locus, downstreamof

^Ca

andincludingalloftheIgH-Vgenes—

replicates early in S, and a transition region no longer through which it is likely that a single replication fork

progresses during most of the S phase. This replication exists. Coupled with additional results of N/A 2D gel electrophoresis in which we detected replication forks fork appears to originate from the last in a cluster of

early activated replicons and proceeds to the first in a moving through IgH-C genes in both directions (O. V. E., L. H. N., and C. L. S., unpublished data), we postulate cluster of late activated replicons (Figure 6).

Inthe buddingyeast,

S. cerevisiae

, asingle replication that at least one additional, B cell–specific origin is acti- vated in the IgH-C locus. In MEL cells, in which this fork has similarly been shown to provide the transition

between early and late replicating regions on chromo- origin(s)issilent,itislikelythattheIgHlocusisreplicated by a single replication fork and forms a transition region some III. An origin is located at ARS 305 near the middle

of the chromosome (Huberman et al., 1988). Initiation between early and late replicated domains. It seems likely, therefore, that the temporal arrangement of DNA occurs early in S, and replication forks then progress

toward the late replicating left telomere (Reynolds et al., replication in mammalian cells is a dynamic develop- mentally regulatedprocess, accompanied bythe activa- 1989; Dubey et al., 1991). In mammalian cells, however,

ourstudiesoftheIgHclusterrepresentauniqueanalysis tion or silencing of origins of DNA replication. Further

Gilbert, and James Borowiec for critical reading of the manuscript.

experiments will determine the mechanisms by which This study is dedicated to the memory of Dr. Ruth M. Kavenoff

silencing or activation of origins occurs. whose exquisite EM images of DNA will continue to enlighten.

Experimental Procedures Received November 10, 1998; revised December 22, 1998.

Cell Culture References

Mouse erythroleukemia cells (MEL) were grown in Dulbecco’s modi-

fied Eagle’s medium, supplemented with 10% fetal bovine serum Aladjem, M.I., Groudine, M., Brody, L.L., Dieken, E.S., Fournier, R.E., (GIBCO BRL), penicillin (100 U/ml), and streptomycin (100

^m

g/ml) at Wahl, G.M., and Epner, E.M. (1995). Participation of the human beta- 37

⁸

C with 5% CO

2

. Cells were maintained as an exponential culture globin locus control region in initiation of DNA replication. Science at a density between 2

³

10

⁵

and 7

³

10

⁵

cells/ml.

₂₇₀

, 815–819.

Aladjem, M.I., Rodewald, L.W., Kolman, J.L., and Wahl, G.M. (1998).

Centrifugal Elutriation Genetic dissection of a mammalian replicator in the human beta- Centrifugal elutriation was performed as described (Brown et al., globin locus. Science

281

, 1005–1009.

1987). Cells were fractionated according to size at 1340 rpm in the

Ariizumi, K., Wang, Z., and Tucker, P.W. (1993). Immunoglobulin JE 10X elutriator rotor of a J6B Beckman centrifuge at 4

⁸

C. The

heavy chain enhancer is located near or in an initiation zone of stage of the cells in S phase was determined by fluorescence-

chromosomal DNA replication. Proc. Natl. Acad. Sci. USA

90

, 3695–

activated cell sorter analysis. Cells in the early and early middle S

3699.

phase fractions were used to prepare matrix-associated DNA.

Bilyeu, K., and Chinault, A.C. (1998). Replication timing properties

DNA Preparation acrossthepseudoautosomalregionboundaryandcytogeneticband

Nuclear matrix-associated DNA was isolated as described (Dijkwel boundaries on human distal Xp. Chromosoma

¹⁰⁷

, 105–112.

et al., 1991). Following dialysis and precipitation, DNA from

z

5

³

Birshtein, B.K., Chen, C., Saleque, S., Michaelson, J.S., Singh, M., 10

⁸

cells was digested with the appropriate restriction enzyme(s) in and Little, R.D. (1997). Murine and human 3

⁹

IgH regulatory se- a volume of 3 ml for 4 hr. This additional restriction digestion was quences. Curr. Top. Microbiol. Immunol.

²²⁴

, 73–80.

necessary since the matrix fraction otherwise showed a significant Brewer, B.J., and Fangman, W.L. (1987). The localization of replica- amount of partial digestion products (our unpublished data). Mouse tion origins on ARS plasmids in

S. cerevisiae

. Cell

⁵¹

, 463–475.

matrix-associated DNA contains a high molecular weight (HMW) Brewer, B.J., and Fangman, W.L. (1988). A replication fork barrier fraction, which is not digested by enzymes used in its preparation. at the 3

⁹

end of yeast ribosomal RNA genes. Cell

⁵⁵

, 637–643.

The HMW DNA (probably the satellite DNA component) binds to the

Brown, E.H., Iqbal, M.A., Stuart, S., Hatton, K.S., Valinsky, J., and BND column and interferes with the chromatography procedure.

Schildkraut, C.L. (1987). Rate of replication of the murine immuno- To separate HMW DNA from the matrix-associated DNA fraction,

globulin heavy-chain locus: evidence that the region is part of a a sucrose gradient is necessary. A 30 ml 10%–40% continuous

single replicon. Mol. Cell. Biol.

⁷

, 450–457.

sucrose gradient in 0.8 M NaCl, 20 mM Tris-HCl (pH 7.4), and 5 mM

EDTA was prepared using a gradient maker. Matrix DNA from 1 to Calvo, C.F., Giannini, S.L., Martinez, N., and Birshtein, B.K. (1991).

5

³

10

⁸

cells in a volume of 0.5 to 5 ml was layered on top of the DNA sequences 3

⁹

of the Ig H chain cluster rearrange in mouse B gradient. The tubes were centrifuged using the Beckman SW28 cell lines. J. Immunol.

146

, 1353–1360.

swinging bucket rotor at 26,000 rpm (83,000 g) at 15

⁸

C for 20–24 Calza,R.E.,Eckhardt,L.,DelGiudice,T.,andSchildkraut,C.L.(1984).

hr. One milliliter fractions were collected by pumping the gradient Changes in gene position are accompanied by a change in time of out from near the bottom of the tube (3 ml remains in the tube as replication. Cell

36

, 689–696.

bottom fraction, and 27 1 ml fractions were collected). Every second Craig, J.M., and Bickmore, W.A. (1993). Chromosome bands—

fraction was analyzed on a 0.5% agarose gel to determine which flavours to savour. Bioessays

15

, 349–354.

fractions contained the DNA fragments in the desired size range. Dhar, V., Mager, D., Iqbal, A., and Schildkraut, C.L. (1988). The coor- The first four to five fractions contained predominantly HMW DNA

dinate replication of the human

^b

-globin gene domain reflects its and were discarded. The desired fractions, which in most experi-

transcriptional activity and nuclease hypersensitivity. Mol. Cell. Biol.

ments were fractions 5–15 (containing fragments from

^,

1 kb up to

8

, 4958–4965.

20–23 kb), were pooled and ethanol precipitated. The entire amount

ofDNArecoveredfromthisprecipitationwasusedforBND-cellulose Dijkwel, P.A., Vaughn, J.P., and Hamlin, J.L. (1991). Mapping of column chromatography, as described (Dijkwel et al., 1991). After replication sites in mammalian genomes by two-dimensional gel precipitation and resuspension, DNA eluted by the caffeine wash analysis: stabilization and enrichment of replication intermediates was quantitated by Hoechst dye binding and examined by N/N or by isolation on the nuclear matrix. Mol. Cell. Biol.

¹¹

, 3850–3859.

N/A 2D gel electrophoresis. Dijkwel,P.A.,Vaughn,J.P.,andHamlin,J.L.(1994).Replicationinitia- tion sites are distributed widely in the amplified CHO dihydrofolate Two-Dimensional Agarose Gel Electrophoresis reductase domain. Nucleic Acids Res.

²²

, 4989–4996.

Approximately 20–24

^m

g DNA from the caffeine wash was subjected Dijkwel, P.A., and Hamlin, J.L. (1996). Sequence and context effects to N/A 2D gel electrophoresis (Little et al., 1993). N/N 2D gel electro- on origin function in mammalian cells. J. Cell. Biochem.

⁶²

, 210–222.

phoresis of fragments between 2 kb and 5 kb was performed as Dubey, D.D., Davis, L.R., Greenfeder, S.A., Ong, L.Y., Zhu, J.G., previously described (Brewer and Fangman, 1988) using 10–14

^m

g Broach, J.R., Newlon, C.S., and Huberman, J.A. (1991). Evidence of caffeine wash DNA. For restriction fragments between 5 and 19 suggesting that the ARS elements associated with silencers of the kb, we used the conditions previously described (Little et al., 1993). yeast mating-type locus HML do not function as chromosomal DNA

replication origins. Mol. Cell. Biol.

11

, 5346–5355.

Southern Transfer and Hybridization Eckhardt,L.A.,andBirshtein,B.K.(1985).Independentimmunoglob- DNA was transferred to Hybond N Plus membranes (Amersham) as ulin class-switch events occurring in a single myeloma cell line. Mol.

recommended by the manufacturer. Cell. Biol.

⁵

, 856–868.

Friedman, K.L., and Brewer, B.J. (1995). Analysis of replication inter-

Acknowledgments mediates by two-dimensional agarose gel electrophoresis. Methods

Enzymol.

²⁶²

, 613–627.

This work was supported by National Institutes of Health grants

Gencheva, M., Anachkova, B., and Russev, G. (1996). Mapping the GM45751 (C. L. S.), AI13509 (B. K. B.), and AI23548 (R. R.). Support sites of initiation of DNA replication in rat and human rRNA genes.

was also provided by Cancer Center Core Grant NIH/NCI P30- J. Biol. Chem.

²⁷¹

, 2608–2614.

CA13330. L. H. N. was supported by a fellowship from the Leukemia

Society of America. We thank Drs. Paolo Norio, Peter Dijkwel, David Gillies, S.D., Morrison, S.L., Oi, V.T., and Tonegawa, S. (1983). A

33

, 717–728. IgH regulatory region includes two virtually iden jor intron

tissue-specific

of a rearranged transcription enhancer immunoglobulin element

heavy is located chain in

gene. the ma-

Cell and Saleque, Birshtein, S., Singh,

B.K. M., (1997). Little,

Dyad R.D., Giannini, symmetry S.L.,

within tical the Michaelson,

mouse enhancers

J.S., 3

⁹

(C alpha3

⁹

E and hs3). J. Immunol.

¹⁵⁸

, 4780–4787.

Goldman, M.A., Holmquist, G.P., Gray, M.C., Caston, L.A., and Nag,

A. (1984). Replication timing of genes and middle repetitive se- Shimizu,A.,Takahashi,N.,Yaoita,Y.,andHonjo,T.(1982).Organiza- tion of the constant-region gene family of the mouse immunoglobu- quences. Science

²²⁴

, 686–692.

lin heavy chain. Cell

²⁸

, 499–506.

Gregor, P.D., and Morrison, S.L. (1986). Myeloma mutant with a

novel 3

⁹

flanking region: loss of normal sequence and insertion Sidow, A., Bulotsky, M.S., Kerrebrock, A.W., Bronson, R.T., Daly, M.J., Reeve, M.P., Hawkins, T.L., Birren, B.W., Jaenisch, R., and of repetitive elements leads to decreased transcription but normal

processing of the alpha heavy-chain gene products. Mol. Biol. Cell Lander,E.S.(1997).Serrate2isdisruptedinthemouselimb-develop- ment mutant syndactylism. Nature

³⁸⁹

, 722–725.

6

, 1903–1916.

Hand, R. (1975). Regulation of DNA replication on subchromosomal Simon, I., and Cedar, H. (1996). Temporal order of DNA replication.

In DNA Replication in Eukaryotic Cells. M.L. DePamphilis, ed. (Cold units of mammalian cells. J. Cell Biol.

⁶⁴

, 89–97.

Spring Harbor, NY: Cold Spring Harbor Laboratory Press), pp.

Hand, R. (1978). Eucaryotic DNA: organization of the genome for 387–408.

replication. Cell

15

, 317–325.

Strehl, S., LaSalle, J.M., and Lalande, M. (1997). High-resolution Hatton, K.S.,and Schildkraut, C.L. (1990).The mouse immunoglobu- analysis of DNA replication domain organization across an R/G- lin kappa light chain genes are located in early and late replicating band boundary. Mol. Cell. Biol.

17

, 6157–6166.

regions of chromosome 6. Mol. Cell. Biol.

10

, 4314–4323.

Tenzen, T., Yamagata, T., Fukagawa, T., Sugaya, K., Ando, A., Inoko, Hatton, K.S., Dhar, V., Brown, E.H., Iqbal, M.A., Stuart, S., Didamo, H., Gojobori, T., Fujiyama, A., Okumura, K., and Ikemura, T. (1997).

V.T., and Schildkraut, C.L. (1988). Replication program of active and

Precise switching of DNA replication timing in the GC content transi- inactive multigene families in mammalian cells. Mol. Cell. Biol.

8

,

tion area in the human major histocompatibility complex. Mol. Cell.

2149–2158. Biol.

¹⁷

, 4043–4050.

Herring, C.D., Chevillard, C., Johnston, S.L., Wettstein, P.J., and Tutter, A., Brodeur, P., Shlomchik, M., and Riblet, R. (1991). Struc- Riblet, R. (1998). Vector-hexamer PCR isolation of all insert ends ture, map position, and evolution of two newly diverged mouse Ig from a YAC contig of the mouse Igh locus. Genome Res.

⁸

, 673–681. VH gene families. J. Immunol.

¹⁴⁷

, 3215–3223.

Huberman,J.A.,andRiggs,A.D.(1966).Autoradiographyofchromo- Yancopoulos,G.D., Desiderio,S.V.,Paskind, M.,Kearney, J.F.,Balti- somal DNA fibers from Chinese hamster cells. Proc. Natl. Acad. Sci. more, D., and Alt, F.W. (1984). Preferential utilization of the most JH-

USA

⁵⁵

, 599–606. proximalVHgenesegmentsinpre–B-celllines.Nature

³¹¹

,727–733.

Huberman, J.A., Zhu, J.G., Davis, L.R., and Newlon, C.S. (1988). Yoon, Y., Sanchez, J.A., Brun, C., and Huberman, J.A. (1995). Map- Close association of a DNA replication origin and an ARS element ping of replication initiation sites in human ribosomal DNA by na- on chromosome III of the yeast, Saccharomyces cerevisiae. Nucleic scent-strand abundance analysis. Mol. Cell. Biol.

¹⁵

, 2482–2489.

Acids Res.

¹⁶

, 6373–6384.

Yurov, Y.B., and Liapunova, N.A. (1977). The units of DNA replication Kitsberg, D., Selig, S., Keshet, I., and Cedar, H. (1993). Replication inthemammalianchromosomes:evidenceforalargesizeofreplica- structure of the human beta-globin gene domain [see comments]. tion units. Chromosoma

60

, 253–267.

Nature

366

, 588–590.

Kobayashi, T., Rein, T., and DePamphilis, M.L. (1998). Identification of primary initiation sites for DNA replication in the hamster dihydro- folate reductase gene initiation zone. Mol. Cell. Biol.

18

, 3266–3277.

Kornberg, A., and Baker, T.A. (1992). Genome Origins. In DNA Repli- cation, A. Kornberg and T.A. Baker, eds. (New York: Freeman W.H.), pp. 511–552.

Lan, Y., Jiang, R., Shawber, C., Weinmaster, G., and Gridley, T.

(1997). The Jagged2 gene maps to chromosome 12 and is a candi- date for the lgl and sm mutations. Mamm. Genome

⁸

, 875–876.

Lang, R.B., Stanton,L.W., and Marcu, K.B.(1982). On immunoglobu- lin heavy chain gene switching: two gamma 2b genes are rearranged via switch sequences in MPC-11 cells but only one is expressed.

Nucleic Acids Res.

¹⁰

, 611–630.

Little, R.D., Platt, T.H.K., and Schildkraut, C.L. (1993). Initiation and termination of DNA replication in human rRNA genes. Mol. Cell. Biol.

13

, 6600–6613.

Michaelson,J.S.,Giannini,S.L.,andBirshtein,B.K.(1995).Identifica- tion of 3

⁹

alpha-hs4, a novel Ig heavy chain enhancer element regu- lated at multiple stages of B cell differentiation. Nucleic Acids Res.

23

, 975–981.

Michaelson,J.S.,Ermakova,O.,Birshtein,B.K.,Ashouian,N.,Chevil- lard, C., Riblet, R., and Schildkraut, C.L. (1997). Regulation of the replication of the murine immunoglobulin heavy chain gene locus:

evaluation of the role of the 3

⁹

regulatory region. Mol. Cell. Biol.

¹⁷

, 6167–6174.

Nawotka, K.A., and Huberman, J.A. (1988). Two-dimensional gel electrophoretic method for mapping DNA replicons. Mol. Cell. Biol.

8

, 1408–1413.

Reynolds, A.E., McCarroll, R.M., Newlon, C.S., and Fangman, W.L.

(1989). Time of replication of ARS elements along yeast chromo- some III. Mol. Cell. Biol.

9

, 4488–4494.

Roeder, W., Maki, R., Traunecker, A., and Tonegawa, S. (1981).

Linkage of the four gamma subclass heavy chain genes. Proc. Natl.

Acad. Sci. USA

⁷⁸

, 474–478.